278177 research outputs found
Sort by
Active Trust Management for Successful Human-Robot Teaming: Moving from a Trust Repair to a Trust Satisficing Perspective
No full textIntegrating mobile robots into human teams promises significant capability improvements for tasks such as searching hazardous environments. Unlike existing teleoperated robots, future robot systems will increasingly be endowed with some level of artificial intelligence (AI), giving them a degree of autonomy in how they pursue mission goals. This autonomy could make a human-agent (robot) team more effective but also put inter-agent trust under strain if robots make a mistake, or (appear to) pursue task priorities that conflict with the team’s best interest. During a mission, agents’ trust states are anticipated to vary according to the situation as understood by each teammate (trustor). If component-level (agent) or system-level trust falls below sufficient levels for cooperative tasks to be completed, it could critically affect mission success . We argue that active trust management will be an important precondition for the success of human-robot teams (HRTs, a subcategory of human-agent teams with embodied agents), especially in dynamic, high-risk environments. We present a trust satisficing perspective which acknowledges and attempts to account for the fluctuating, multi-faceted, and context-dependent nature of trust and trust requirements even under normal operating conditions. Our outline of a trust management framework for human-robot teaming includes online measurement of proxy metrics for trust, closed-loop adaptation of robot behavior, and variable autonomy to give space for human responsibility in situations requiring value judgements. We refer to a recent experimental exploration of ‘swift trust’ and a novel behavioral trust metric for HRT, and we highlight issues for further investigation
Teaching comparative law post-Brexit and post-SQE:Challenges and choices for UK universities and teachers of comparative law
No full textThis article will reflect on the impact of Brexit and of the introduction of the SQE on comparative law teaching in UK universities. Drawing on a 2025 survey by the British Association of Comparative Law (BACL) on how comparative law is currently taught in UK universities, it will examine the place of comparative law in the Law School curriculum. In so doing, it will identify changes since the last BACL teaching survey of 2002. Has Brexit, with EU law potentially demoted to an optional part of the Law degree curriculum, discouraged interest in studying comparative law? What has been the impact of the introduction of the SQE on Law School curricula? Having been given exclusive access to the findings of the 2025 BACL teaching survey, I will examine the challenges and choices universities and comparative law academics face. Is comparative law teaching an interesting but ornamental addition to a crammed curriculum or an important part of the UK university response to globalisation and the internationalisation of legal education
New Control Functionalities for Launcher Load Relief in Ascent and Descent Flight
Get PDFThe development of effective load relief strategies is key to the improvement of launcher flight performance as it enables a joint increase of wind resilience and decrease of mass. This is particularly relevant for reusable launchers, which are aimed at maximising their operational availability and payload capacity. Yet, despite various advances in aeronautics and wind energy load relief, classical feedback-only techniques remain the state-of-practice for launchers. In this article, an improved load relief functionality is proposed by augmenting a conventional control design with a disturbance observer for on-board wind anticipation. Without requiring any change in the feedback loop,this approach also paves the way for the use of forward looking wind estimation in the launcher domain. The disturbance observer is designed and analysed using robust control techniques for a lightweight, non-winged reusable launch vehicle, which relies on the use of planar fins for unpropelled descent attitude control. While the use of fins for launch is not a common practice, the article also exemplifies how their use can further improve launcher performance
Class, Race and Lifestyles in the US:A Play of Spaces
No full textPierre Bourdieu’s famous thesis on the homology between class and lifestyles has been accused of overlooking or downplaying race and ethnicity. The most sophisticated tests and updates of his thesis in statistical research have so far done little to counteract this, even though recent advances in theorising ethno-racial domination and its intersection with class have been inspired by Bourdieu’s field theory. This paper attempts to bring the divergent threads of scholarship together by conceiving and exploring the relationship between class, race and lifestyles in terms of intersecting spaces. Using survey data from the US and multiple correspondence analysis, I build models of the US ‘social space’ and lifestyle space and examine the position and dispersion of ethno-racial categories in both. I then use regression analysis to document the relative autonomy of race in differentiating lifestyles and where in the class structure it makes the most difference. While race has a cross-class effect on lifestyles, this is nonetheless modulated by capital holdings. For some groups, money is a more accessible route to dominant culture than Eurocentric highbrow taste and consumption
Reasoning That Leaks, Fine-Tuning That Amplifies:Exposing the Hidden Threats of Chain-of-Thought Models
No full textChain-of-Thought (CoT) guides large language models to reason step-by-step, yielding remarkable performance gains across diverse tasks. However, this structured reasoning process also introduces novel and underexplored security risks. In this paper, we present an in-depth analysis of fine-tuning attacks targeting CoT-enabled LLMs, with particular focus on "aha moments" during reasoning, which are critical intermediate steps the model takes to make a significant decision or change its behavior. Through experiments on six CoT models and three non-CoT baselines, we find that even aligned CoT models can be more harmful than their base models. Moreover, the reasoning process frequently contains more harmful and actionable content than the final answer, even when the final answer refuses a harmful request. By examining the causal relationship between the reasoning process and the final outputs, we identify two distinct failure modes, Unintentional Leakage and Harmful Escalation, that systematically drive the generation of harmful reasoning. To rigorously assess these risks, we propose an evaluation framework grounded in the EU AI Act and construct a policy-aligned benchmark dataset for CoT reasoning. Our findings expose inherent vulnerabilities in CoT and offer insights for supervising and aligning the reasoning process in LLMs
Pile driving installation effects in low-to-medium density chalk
No full textThis paper explores the effects of pile driving installation on low-to-medium density chalk. The damage developed around eight open and closed, steel and concrete, piles driven at the St Nicholas at Wade test site in Kent, UK, was investigated by carefully logging and micro-sampling chalk from around their shafts after long ageing in situ. The observations identified relatively thin annuli of de-structured 'putty' chalk left around the pile shafts. Related numerical studies have confirmed that, after re-consolidation in situ, the Zone A material largely controls the piles’ axial load-displacement behaviour. Measurements of the reduced Zone A water content profiles around piles with different geometries, scales and materials confirm that their thicknesses scale primarily with pile wall thickness tw. A second, more extensive, annular Zone B was also identified, which manifests far more frequent fracturing than the natural chalk. Earlier numerical analyses have shown that its degraded properties largely control the piles’ lateral loading responses. The Zone B annular thicknesses, and degrees of damage within them, depend on both pile diameter D and tw. These observations are crucial to both modelling piles driven in chalk and any lateral loading design calculations
The Gulag Letters:The Soviet History of Gay Love
No full textThis essay complements Dan Healey’s research both in its use of new sources and on the conceptual level by reconstructing the story of a male couple, two composers, who lived together during the 1920s, a relatively liberal period for gay people in the USSR, and were separated in the 1930s following the reintroduction of the criminal article against homosexuality. Their previously unresearched extensive correspondence, written while one of them was in the Gulag and the other remained in Moscow, forms the core source base of this study, adding an intimate and profoundly human dimension to the history of homosexuality in the USSR. It shows how one of the first men charged under the article criminalizing homosexuality survived in the Gulag while maintaining relationship with a partner on the outside
Structured tools for assessing quality and risk of bias in Mendelian randomisation studies:an updated systematic review
No full textBackground: The growing use of Mendelian randomisation (MR) has heightened the need for rigorous quality and bias assessment tools. A previous systematic review included studies published up to July 2021 identified 14 structured instruments for conducting, evaluating, and reporting MR studies. However, methodological developments have accelerated in the years since. Methods: We updated the previous systematic review to include tools published between July 2021 and January 2025, applying the same search strategy and eligibility criteria. Two reviewers independently screened articles, extracted data, and mapped tool content to bias domains. Results: We identified 15 additional articles, bringing the total to 29 tools. Of these 29, 19 provided structured evaluation tools. 12 of the 19 evaluation tools were newly added in the present review, which addressed broader methodological domains beyond core instrumental variable assumptions, including genetic instrument selection, population stratification, sensitivity analyses, and dataset considerations. However, substantial variation in bias domains, structure, and scoring methods across tools persists. Key gaps remain in the assessment of linkage disequilibrium, missing data, and dynastic effects. Conclusions: While the number of structured tools has increased in recent years, the lack of standardisation across tools makes it difficult to compare across systematic reviews of MR studies. Developing more complete and standardised evaluation frameworks and properly testing these tools in practice are important next steps to improve the overall quality of MR research
Blind Deduction and Syntax
No full textLuo, Horsten, and Roberts (2024) recently responded to a version of the so-called conservativeness argument against deflationism about truth that I presented in (Fujimoto 2022). This article replies to them and raises two issues with their argument. However, despite these issues, their argument and the disagreements between them and me bring a new perspective to the fundamental question as to what the theory of syntax ought to be. In the final section, I explore this question and propose one possible argument that any adequate theory of self-applicable truth should derive the full schema of arithmetical induction for the language of theories of truth over arithmetic